CRIN Logo - White Stacked

    Machine Learning for False Positive Elimination in Continuous Methane Monitoring with Low-Cost IR Cameras

    Solution Developer

    Kuva Systems Inc.

    Project Description

    Kuva’s industrial IoT solution continuously monitors and quantifies methane and VOC emissions, providing actionable alerts with no false positive readings.

    Methane emissions from the oil and gas industry are a major contributor to climate change. Recent studies indicate the majority of upstream emissions result from intermittent and abnormal operations, especially from sites with tanks. Visual continuous monitoring is needed to detect and fix problems fast.

    The Kuva infrared imaging system automatically detects and measures emissions, delivering direct image-based alerts. Those images of emissions, annotated with quantified release data, are then transferred via the cloud to customers’ work order management and production operation systems. Armed with this actionable information, customers can implement mitigation plans without the need to conduct secondary manual inspections.

    The Kuva solution consists of three elements: the Kuva camera based on short-wave infrared (SWIR) imaging, a cloud solution for further data processing and leak rate quantification, and a monitoring and notification service that includes a final review to eliminate any remaining false positive readings before the images are transferred to oil and gas customers.

    Benefits/Outcomes

    • Visualization and notification of emission events with no false positives

    • Enables timely investigation and fix of the root cause of emissions

    • Scalable and affordable continuous methane and VOC monitoring

    Project Resources

    http://www.kuvasystems.com/

    Project Video

    Maximum Funding from CRIN

    $760,605

    Main Project Contact

    Monica Sippola

    msippola@kuvasystems.com

    Technology Readiness Level

    TRL 9 - Actual technology, product and/or process proven through successful deployment in an operational setting